1. Field of the Invention
The present invention is related to the fields of vaccinology, immunology and medicine. The invention provides compositions and methods for enhancing immunological responses against virus-like particles (VLPs) or against antigens coupled, fused or attached otherwise to VLPs by packaging immunostimulatory substances, in particular immunostimulatory nucleic acids, and even more particular oligonucleotides containing at least one non-methylated CpG sequence, into the VLPs. The invention can be used to induce strong and sustained T cell responses particularly useful for the treatment of tumors and chronic viral diseases as well as allergies and other chronic diseases.
2. Related Art
The essence of the immune system is built on two separate foundation pillars: one is specific or adaptive immunity which is characterized by relatively slow response-kinetics and the ability to remember; the other is non-specific or innate immunity exhibiting rapid response-kinetics but lacking memory. Lymphocytes are the key players of the adaptive immune system. Each lymphocyte expresses antigen-receptors of unique specificity. Upon recognizing an antigen via the receptor, lymphocytes proliferate and develop effector function. Few lymphocytes exhibit specificity for a given antigen or pathogen, and massive proliferation is usually required before an effector response can be measured—hence, the slow kinetics of the adaptive immune system. Since a significant proportion of the expanded lymphocytes survive and may maintain some effector function following elimination of the antigen, the adaptive immune system reacts faster when encountering the antigen a second time. This is the basis of its ability to remember.
In contrast to the situation with lymphocytes, where specificity for a pathogen is confined to few cells that must expand to gain function, the cells and molecules of the innate immune system are usually present in massive numbers and recognize a limited number of invariant features associated with pathogens (Medzhitov, R. and Janeway, C. A., Jr., Cell 91:295-298 (1997)). Examples of such patterns include lipopolysaccharides (LPS), non-methylated CG-rich DNA (CpG) or double stranded RNA, which are specific for bacterial and viral infections, respectively.
Most research in immunology has focused on the adaptive immune system and only recently has the innate immune system entered the focus of interest. Historically, the adaptive and innate immune system were treated and analyzed as two separate entities that had little in common. Such was the disparity that few researchers wondered why antigens were much more immunogenic for the specific immune system when applied with adjuvants that stimulated innate immunity (Sotomayor, E. M., et al., Nat. Med. 5:780 (1999); Diehl, L., et al., Nat. Med. 5:774 (1999); Weigle, W. O., Adv. Immunol. 30:159 (1980)). However, the answer posed by this question is critical to the understanding of the immune system and for comprehending the balance between protective immunity and autoimmunity.
Rationalized manipulation of the innate immune system and in particular activation of APCs involved in T cell priming to deliberately induce a self-specific T cell response provides a means for T cell-based tumor-therapy. Accordingly, the focus of most current therapies is on the use of activated dendritic cells (DCs) as antigen-carriers for the induction of sustained T cell responses (Nestle et al., Nat. Med. 4:328 (1998)). Similarly, in vivo activators of the innate immune system, such as CpGs or anti-CD40 antibodies, are applied together with tumor cells in order to enhance their immunogenicity (Sotomayor, E. M., et al., Nat. Med. 5:780 (1999); Diehl, L., et al., Nat. Med. 5:774 (1999)).
Generalized activation of APCs by factors that stimulate innate immunity may often be the cause for triggering self-specific lymphocytes and autoimmunity. Activation may result in enhanced expression of costimulatory molecules or cytokines such as IL-12 or IFNα. This view is compatible with the observation that administration of LPS together with thyroid extracts is able to overcome tolerance and trigger autoimmune thyroiditis (Weigle, W. O., Adv. Immunol. 30:159 (1980)). Moreover, in a transgenic mouse model, it was recently shown that administration of self-peptide alone failed to cause auto-immunity unless APCs were activated by a separate pathway (Garza, K. M., et al., J. Exp. Med. 191:2021 (2000)). The link between innate immunity and autoimmune disease is further underscored by the observation that LPS, viral infections or generalized activation of APCs delays or prevents the establishment of peripheral tolerance (Vella, A. T., et al., Immunity 2:261 (1995); Ehl, S., et al., J. Exp. Med. 187:763 (1998); Maxwell, J. R., et al., J. Immunol. 162:2024 (1999)). In this way, innate immunity not only enhances the activation of self-specific lymphocytes but also inhibits their subsequent elimination. These findings may extend to tumor biology and the control of chronic viral diseases.
Induction of cytotoxic T lymphocyte (CTL) responses after immunization with minor histocompatibility antigens, such as the HY-antigen, requires the presence of T helper cells (Th cells) (Husmann, L. A., and M. J. Bevan, Ann. NY. Acad. Sci. 532:158 (1988); Guerder, S., and P. Matzinger, J. Exp. Med. 176:553 (1992)). CTL-responses induced by cross-priming, i.e. by priming with exogenous antigens that reached the class I pathway, have also been shown to require the presence of Th cells (Bennett, S. R. M., et al., J. Exp. Med. 186:65 (1997)). These observations have important consequences for tumor therapy where T help may be critical for the induction of protective CTL responses by tumor cells (Ossendorp, F., et al., J. Exp. Med. 187:693 (1998)).
An important effector molecule on activated Th cells is the CD40-ligand (CD40L) interacting with CD40 on B cells, macrophages and dendritic cells (DCs) (Foy, T. M., et al., Annu. Rev. Immunol. 14:591 (1996)). Triggering of CD40 on B cells is essential for isotype switching and the generation of B cell memory (Foy, T. M., et al., Ann. Rev. Immunol. 14:591 (1996)). More recently, it was shown that stimulation of CD40 on macrophages and DCs leads to their activation and maturation (Cella, M., et al., Curr. Opin. Immunol. 9:10 (1997); Banchereau, J., and R. M. Steinman Nature 392:245 (1998)). Specifically, DCs upregulate costimulatory molecules and produce cytokines such as IL-12 upon activation. Interestingly, this CD40L-mediated maturation of DCs seems to be responsible for the helper effect on CTL responses. In fact, it has recently been shown that CD40-triggering by Th cells renders DCs able to initiate a CTL-response (Ridge, J. P., et al., Nature 393:474 (1998); Bennett, S. R. M., et al., Nature 393:478 (1998); Schoenenberger, S. P., et al., Nature 393:480 (1998)). This is consistent with the earlier observation that Th cells have to recognize their ligands on the same APC as the CTLs, indicating that a cognate interaction is required (Bennett, S. R. M., et al., J. Exp. Med. 186:65 (1997)). Thus CD40L-mediated stimulation by Th cells leads to the activation of DCs, which subsequently are able to prime CTL-responses. In the human, type I interferons, in particular interferon α and β may be equally important as IL-12.
In contrast to these Th-dependent CTL responses, viruses are often able to induce protective CTL-responses in the absence of T help (for review, see (Bachmann, M. F., et al., J. Immunol. 161:5791 (1998)). Specifically, lymphocytic choriomeningitis virus (LCMV) (Leist, T. P., et al., J. Immunol. 138:2278 (1987); Ahmed, R., et al., J. Virol. 62:2102 (1988); Battegay, M., et al., Cell Immunol. 167:115 (1996); Borrow, P., et al., J. Exp. Med. 183:2129 (1996); Whitmire, J. K., et al., J. Virol. 70:8375 (1996)), vesicular stomatitis virus (VSV) (Kündig, T. M., et al., Immunity 5:41 (1996)), influenza virus (Tripp, R. A., et al., J. Immunol. 155:2955 (1995)), vaccinia virus (Leist, T. P., et al., Scand. J. Immunol. 30:679 (1989)) and ectromelia virus (Buller, R., et al., Nature 328:77 (1987)) were able to prime CTL-responses in mice depleted of CD4+ T cells or deficient for the expression of class II or CD40. The mechanism for this Th cell independent CTL-priming by viruses is presently not understood. Moreover, most viruses do not stimulate completely Th cell independent CTL-responses, but virus-specific CTL-activity is reduced in Th-cell deficient mice. Thus, Th cells may enhance anti-viral CTL-responses but the mechanism of this help is not fully understood yet. DCs have recently been shown to present influenza derived antigens by cross-priming (Albert, M. L., et al., J. Exp. Med. 188:1359 (1998); Albert, M. L., et al., Nature 392:86 (1998)). It is therefore possible that, similarly as shown for minor histocompatibility antigens and tumor antigens (Ridge, J. P., et al., Nature 393:474 (1998); Bennett, S. R. M., et al., Nature 393:478 (1998); Schoenenberger, S. P., et al., Nature 393:480 (1998)), Th cells may assist induction of CTLs via CD40 triggering on DCs. Thus, stimulation of CD40 using CD40L or anti-CD40 antibodies may enhance CTL induction after stimulation with viruses or tumor cells.
However, although CD40L is an important activator of DCs, there seem to be additional molecules that can stimulate maturation and activation of DCs during immune responses. In fact, CD40 is not measurably involved in the induction of CTLs specific for LCMV or VSV (Ruedl, C., et al., J. Exp. Med. 189:1875 (1999)). Thus, although VSV-specific CTL responses are partly dependent upon the presence of CD4+ T cells (Kündig, T. M., et al., Immunity 5:41 (1996)), this helper effect is not mediated by CD40L. Candidates for effector molecules triggering maturation of DCs during immune responses include Trance and TNFα (Bachmann, M. F., et al., J. Exp. Med. 189:1025 (1999); Sallusto, F., and A. Lanzavecchia, J Exp Med 179:1109 (1994)).
It is well established that the administration of purified proteins alone is usually not sufficient to elicit a strong immune response; isolated antigen generally must be given together with helper substances called adjuvants. Within these adjuvants, the administered antigen is protected against rapid degradation, and the adjuvant provides an extended release of a low level of antigen.
Unlike isolated proteins, viruses induce prompt and efficient immune responses in the absence of any adjuvants both with and without T-cell help (Bachmann & Zinkernagel, Ann. Rev. Immunol. 15:235-270 (1997)). Although viruses often consist of few proteins, they are able to trigger much stronger immune responses than their isolated components. For B cell responses, it is known that one crucial factor for the immunogenicity of viruses is the repetitiveness and order of surface epitopes. Many viruses exhibit a quasi-crystalline surface that displays a regular array of epitopes which efficiently crosslinks epitope-specific immunoglobulins on B cells (Bachmann & Zinkernagel, Immunol. Today 17:553-558 (1996)). This crosslinking of surface immunoglobulins on B cells is a strong activation signal that directly induces cell-cycle progression and the production of IgM antibodies. Further, such triggered B cells are able to activate T helper cells, which in turn induce a switch from IgM to IgG antibody production in B cells and the generation of long-lived B cell memory—the goal of any vaccination (Bachmann & Zinkernagel, Ann. Rev. Immunol. 15:235-270 (1997)). Viral structure is even linked to the generation of anti-antibodies in autoimmune disease and as a part of the natural response to pathogens (see Fehr, T., et al., J. Exp. Med. 185:1785-1792 (1997)). Thus, antigens on viral particles that are organized in an ordered and repetitive array are highly immunogenic since they can directly activate B cells.
In addition to strong B cell responses, viral particles are also able to induce the generation of a cytotoxic T cell response, another crucial arm of the immune system. These cytotoxic T cells are particularly important for the elimination of non-cytopathic viruses such as HIV or Hepatitis B virus and for the eradication of tumors. Cytotoxic T cells do not recognize native antigens but rather recognize their degradation products in association with MHC class I molecules (Townsend & Bodmer, Ann. Rev. Immunol. 7:601-624 (1989)). Macrophages and dendritic cells are able to take up and process exogenous viral particles (but not their soluble, isolated components) and present the generated degradation product to cytotoxic T cells, leading to their activation and proliferation (Kovacsovics-Bankowski et al., Proc. Natl. Acad. Sci. USA 90:4942-4946 (1993); Bachmann et al., Eur. J. Immunol. 26:2595-2600 (1996)).
Viral particles as antigens exhibit two advantages over their isolated components: (1) due to their highly repetitive surface structure, they are able to directly activate B cells, leading to high antibody titers and long-lasting B cell memory; and (2) viral particles, but not soluble proteins, have the potential to induce a cytotoxic T cell response, even if the viruses are non-infectious and adjuvants are absent.
Several new vaccine strategies exploit the inherent immunogenicity of viruses. Some of these approaches focus on the particulate nature of the virus particle; for example see Harding, C. V. and Song, R., (J. Immunology 153:4925 (1994)), which discloses a vaccine consisting of latex beads and antigen; Kovacsovics-Bankowski, M., et al. (Proc. Natl. Acad. Sci. USA 90:4942-4946 (1993)), which discloses a vaccine consisting of iron oxide beads and antigen; U.S. Pat. No. 5,334,394 to Kossovsky, N., et al., which discloses core particles coated with antigen; U.S. Pat. No. 5,871,747, which discloses synthetic polymer particles carrying on the surface one or more proteins covalently bonded thereto; and a core particle with a non-covalently bound coating, which at least partially covers the surface of said core particle, and at least one biologically active agent in contact with said coated core particle (see, e.g., WO 94/15585).
In a further development, virus-like particles (VLPs) are being exploited in the area of vaccine production because of both their structural properties and their non-infectious nature. VLPs are supermolecular structures built in a symmetric manner from many protein molecules of one or more types. They lack the viral genome and, therefore, are noninfectious. VLPs can often be produced in large quantities by heterologous expression and can be easily be purified.
In addition, DNA rich in non-methylated CG motifs (CpG), as present in bacteria and most non-vertebrates, exhibits a potent stimulatory activity on B cells, dendritic cells and other APC's in vitro as well as in vivo. Although bacterial DNA is immunostimulatory across many vertebrate species, the individual CpG motifs may differ. In fact, CpG motifs that stimulate mouse immune cells may not necessarily stimulate human immune cells and vice versa.
Although DNA oligomers rich in CpG motifs can exhibit immunostimulatory capacity, their efficiency is often limited, since they are unstable in vitro and in vivo. Thus, they exhibit unfavorable pharmacokinetics. In order to render CpG-oligonucleotides more potent, it is therefore usually necessary to stabilize them by introducing phosphorothioate modifications of the phosphate backbone.
A second limitation for the use of CpG-oligonucleotides to stimulate immune responses is their lack of specificity, since all APC's and B cells in contact with CpG-oligonucleotides become stimulated. Thus, the efficiency and specificity of CpG-oligonucleotides may be improved by stabilizing them or packaging them in a way that restricts cellular activation to those cells that also present the relevant antigen.
In addition, immunostimulatory CpG-oligodeoxynucleotides induce strong side effects by causing extramedullary hemopoiesis accomponied by splenomegaly and lymphadenopathy in mice (Sparwasser et al., J. Immunol. (1999), 162:2368-74 and Example 18).
VLPs have been shown to be efficiently presented on MHC class I molecules as they, presumably after uptake by macropinocytosis, are efficiently processed and crossprimed onto MHC class I. The mechanism of crosspriming is not clear to date, but TAP-dependent and TAP-independent pathways have been proposed.
There have been remarkable advances made in vaccination strategies recently, yet there remains a need for improvement on existing strategies. In particular, there remains a need in the art for the development of new and improved vaccines that promote a strong CTL immune response and anti-pathogenic protection as efficiently as natural pathogens in the absence of generalized activation of APCs and other cells.